How To Drive Change Safely

First, we must understand that in our planning we want to leave the underlying scientific process of our experiments untouched.

In other words, the goal is to drive change where it doesn’t affect our science.

We optimize the way in which we handle the objects that help us assess our samples. We don’t want to alter anything that links to the fundamental experimental principles. Meaning we might change the gradient of our HPLC but the elution principle and resolution stay the same. The two thin lines connecting Handling and Measuring should indicate that sometimes we have to handle our items in a very specific way, as they would otherwise influence the system we measure.

This is why sustainability-related change doesn’t mean significant investments of additional time, effort, or risk.

The biggest reason we are afraid of change is that we prefer to keep things steady because we are already busy with everything else.

Understanding Perceived Risk

We rarely learn how to optimize, how to drive change, and sometimes we don’t even know why our protocols work.

Refer back to our ​lesson on experimental design​, where we discussed sample sizes, statistical planning, and examined an example of how to implement step-wise designs, given the time savings for scientists and the resource savings for the lab.

This makes change feel overwhelming. It makes our mind rationalize irrational concerns.

Generally, it is often possible to find a technical solution – the difficult part is clearly expressing the kind of anxiety that holds us back.

While rationalization was first characterized by psychoanalysis, I found this very well-designed graphic in a ​publication by Cushman​ titled fittingly: “Rationalization is Rational.” Exactly this makes it so difficult for us scientists to realize that we are searching for reasons to support our habits instead of analyzing the data and odds in the present situation before deciding.

To allow us to think accurately, we need to identify and label the anxiety that we are often not aware of.

> Otherwise, we see dangers everywhere or nowhere. But we want to see the actual risks and address them.

Let us therefore look at some examples.

  • A common perceived risk is overlooking a danger and thereby causing contamination. However, if we plan our optimization ahead and review it on two different days, it is very unlikely that we would make such a blunder.
  • Secondly, one might be afraid of being distracted and thereby making mistakes. For example, when reusing the same pipette tip and causing contamination. While this can always happen (why else would cell cultures get contaminated), one might even argue that because we work more sustainably, we pay more attention to best practices and focus more on our experiments.
Mistakes happen to all of us … but the point about anxiety, for example, when it comes to contamination, is simply related to human error. Either you can reuse your tip or you cannot. Keeping the tip and pipetting into the wrong well is a human mistake. And don’t rationalize, you won’t truly “get used to” keeping your tip or serological pipette, as the cases where reuse is possible are limited.
  • Also, some fear that after a change the experiment no longer works as it did previously. But if this were the case, we can simply return to the original method. Of note, investigating surprising failures has been the starting point for ​breakthrough​ findings.

The True Danger

This means most dangers introduced to scientific processes are linked to human failure, not the riskiness of optimization.

In other words, we (rightfully) perceive excessive and therefore blinding eagerness to save resources as the danger.

However, this eagerness fades once we understand it’s not about reaching a zero footprint.

Big changes like exchanging plastic for glass do not make the biggest impact. Reduction does. These graphs come from ​Farley & Benoit​. In short, they tried to extrapolate the footprint of single-use plastic vs. reused plastic vs. glass and found that, in some scenarios, reusing plastic can be as sustainable, if not more sustainable, if we look at CO₂e.

Many scientists fear that sustainability will fundamentally change their science because they assume we must radically cut resources. This is not the case. The goal is optimization.

This is also why radical changes are often not the most sustainable – for example, switching all plasticware to glass is not necessarily efficient, because glass has a large footprint itself, sometimes greater than reused plastics.

My goal is to make you see sustainability-driven changes as a synonym for optimization.

We essentially optimize; we don’t change. And that also means it’s not primarily about the environment – it is about optimizing workflows.

The point is that saving time, chemicals, or plastics naturally translates into sustainability.

The Levels We Operate On

This means most dangers introduced to scientific processes are linked to human failure, not the riskiness of optimization.

Unless we choose to use innovative technology (e.g., switching from conventional assays to SPR), the underlying methodology always stays the same.

That means we might, for example, change the items that we use—this is a change in procurement. We might purchase bio-based materials that are of the same quality.

We may change how we handle our items. For instance, instead of using a 50 mL tube, we use a 15 mL tube. We may also change how we handle our instruments—for example, turning them off during the weekend or optimizing their settings, such as when we use a microscope.

Finally, we can choose to optimize our experimental design—such as when working with mice—so that in the end we have a different setup, for example by changing sample numbers or investigation schedules.

However, none of these require us to change fundamentals of our protocols or rely on assumptions about biological or chemical processes we cannot be certain about.

In fact, protocol optimization, when done as a team, doesn’t require senior expertise from every member. A bachelor’s student can make a suggestion that is valued by a postdoc, as I have experienced myself. In other words, experience should validate proposed changes, but optimization simply requires a thorough understanding of the underlying science.

How It Benefits Science

In essence, we can reframe the question of sustainability and safety.

Sustainability efforts should remove unnecessary steps—steps that are often inherited from outdated purposes because protocols were not optimized or were reused for something else.

This can have significant effects because, in many cases, we have not tapped into the potential for optimization at all.

In many academic settings, protocols were never optimized, and thus adjustments can make huge differences: reductions of 30–50% in solvent use, 60–80% in plastic waste, and sometimes cutting analysis time in half.

Even in industry, where protocols tend to be more optimized, these optimizations usually focus on the scientific aspects, not the handling of items – an area that represents a major opportunity for cost savings.

My goal is to make you see sustainability-driven changes as a synonym for optimization. We essentially optimize; we don’t change.

And that also means it’s not primarily about the environment—it is about optimizing workflows. Saving time, chemicals, or plastics naturally translates into sustainability.

Once we adopt this perspective, we also see the scientific advantages.

We often assume that our experiments measure our target processes without noticing how unnecessary manipulations alter the system.

Sustainability removes unnecessary steps—such as diluting solutions in intermediate tubes—thereby reducing opportunities for contamination or error. In other words, reducing unnecessary steps improves scientific quality.

How To Do It

To implement change safely, following five core principles might help:

  • Differentiation There is no single strategy that works everywhere. Reusing a pipette tip might be appropriate for certain controls where only the concentration differs but the analyte is identical, whereas it would not be appropriate for specific sample types. We should avoid oversimplification.
  • Stepwise implementation Many protocols offer multiple points for optimization. Although it may be tempting, the best approach is stepwise change. This reduces cognitive burden and ensures that if unexpected difficulties arise, we can trace them back and handle them on the spot.
  • Mindset When implementing change for the first time, we need the right mindset. This means planning and analyzing the change beforehand so that we can remain fully present when conducting the experiment. It also means working in a good flow—not being distracted by worries about colleagues’ reactions or anxiety about the change itself. Confidence, focus, and concentration are essential.
  • Experience We must be sufficiently familiar with the protocol before optimizing it. Protocols that are handed down should first be learned and implemented as they are. Then, optimizing includes talking to lab colleagues about potential difficulties and thoroughly reviewing literature to see whether similar changes have been reported.
Protocols can be long and nuanced. Make sure you understand why you do each step and that you remember it sufficiently well to be able to focus on the changes. We often underestimate how quickly we forget – so making a plan and believing you don’t need to take notes or that you can implement changes on the fly is not a good idea. As shown in the diagram about ​spaced repetition​, we generally want to have the original protocol properly established before making changes.
  • Controls This involves performing trial runs to verify whether optimizations are still valid. Then, it’s about documenting changes through controls – in cell culture, for example, this means checking whether cells grow with the same morphology, speed, and metabolism when a new dish type or dish-reuse strategy is employed.

Applying The Knowledge

The key is not to change a running system in the middle of an experiment.

Instead, change should be implemented after an experimental series is completed or when a new project is started.

As we discussed, ​​seven funding bodies ​​met in Heidelberg to support these statements, companies support you with ​​innovations​​, ​​data​​, or ​​tools​​, and initiatives like those by ​​My Green Lab ​​show that pharma companies regularly achieve a positive ROI.

Most often, optimization is missing due to psychological barriers.

One such barrier is insufficient trust in oneself.

A very different one is the reluctance to accept that improvements already exist, as this would mean admitting that one could have been more efficient for a long time or that someone else might find a solution one did not.

When leaders raise doubts, resistance might stems from distrust in a person, not distrust in the change itself (although expressed as such).

However, if you as a supervisor doubt a person’s ability to optimize, you might ask yourself whether you truly trust that person to conduct experiments properly – and if not, why. Then, of course, it is your responsibility to grow them or remove them. Otherwise, what does this imply for the future of the project or the group

Nevertheless, if doubts remain after meticulous planning, change should be aborted. This may indicate that the optimization has the wrong target, impacting aspects of the underlying process.

It might also suggest that the protocol itself is so fragile that it yields essentially artificial data, which requires you to make a call: fundamentally rework the protocol or live with it.

As scientists, we constantly work on new projects and approaches, meaning we are inherently used to change. However, the circumstances matter and this is what we need to express.

If you want to convince yourself, colleagues, or supervisors:

1. Adhere to best practices: research the literature, plan the change carefully, and run a trial.

There are several publications nowadays, whether in ​microbiology​, ​biochemistry​, ​analytical​​chemistry​, or ​synthesis​. Read them to understand how change can be realized. Moreover, they are great for convincing others that change is safely possible.

2. Assure superiors that you will invest the extra time required to avoid losses in productivity, even though long-term benefits are almost always observed.

3. Prove that you are capable of managing the change by clearly articulating what we know and plan to do, even if it feels obvious or trivial because others don’t know what you do.

Once again, the biggest challenge is not the science. it’s the psychological barriers we are rarely aware of.

References

Penndorf, P., et al., 2023. A new approach to making scientific research more efficient – rethinking sustainability. FEBS Letters, 597(19), pp.2371–2374. doi:10.1002/1873-3468.14736.

Farley, M., et al., 2023. Re-use of laboratory utensils reduces CO₂ equivalent footprint and running costs. PLOS ONE, 18(4), e0283697. doi:10.1371/journal.pone.0283697.

Cushman, F., et al., 2020. Rationalization is rational. Behavioral and Brain Sciences, 43, e28. doi:10.1017/S0140525X19001730.

Alves, J., et al., 2020. A case report: insights into reducing plastic waste in a microbiology laboratory. Access Microbiology, 3(3), 000173. doi:10.1099/acmi.0.000173.

Kilcoyne, J., et al., 2022. Reducing environmental impacts of marine biotoxin monitoring: A laboratory report. PLOS Sustainability and Transformation, 1(3), e0000001 doi:10.1371/journal.pstr.0000001.

Mazzali, D., et al., 2025. Sustainable and surfactant-free synthesis of negatively charged acrylamide nanogels for biomedical applications. Macromolecules, 58(3), pp.1206–1213. doi:10.1021/acs.macromol.4c02128.